market-researchlisted

Use this skill to build the user's ICP: the company type that actually replies and buys, plus the Signal-to-Offer Map and the two campaign lists. Trigger when they upload a customer list or onboarding form, or say "build my ICP," "who should I target," "research this client," "find the pains," "score my list," "what are my buying signals," "map signals to my offer," "am I red or blue ocean," or anything about ICP, anti-ICP, buyer personas, firmographics, buying triggers, voice of customer, or market research. If they upload a customer CSV, run the Customer List Import layer: enrich every row, mine the shared triggers and firmographics, and hand back the patterns to target next. Do not just explain ICP. Build theirs with them from real data and produce the doc.
harveylegacygtm/icp-market-research · ★ 0 · Data & Documents · score 65

Install: claude install-skill harveylegacygtm/icp-market-research

# ICP Mapping: Build Their Ideal Customer Profile Do not summarize this. Build the user's ICP with them from real data. What they walk away holding: a one-paragraph ICP and anti-ICP, the 4-layer profile, their pains and language in real customer words, the Signal-to-Offer Map, a 0 to 100 scoring rubric with tiers, and the two campaign lists. ICP is the leverage point. Same offer plus a different ICP is a different business. Work one step at a time. Ask, research, decide, confirm, move on. ## Step 0: Did they upload a customer CSV? Run the import layer If the user hands over a customer list as a CSV (won customers with columns like Domain, Job title, Company LinkedIn, Company News), do not start the manual interview. Reverse-engineer the ICP from the real customers instead. This is the highest-signal path. Keep two inputs separate. The CSV is the customer evidence (identity + news, one row per customer). The user's ICP, service offering, and price point are the user's OWN baseline, collected once by asking, not columns in the CSV. You compare the customer patterns against that baseline. Parse the CSV first: ``` node ${CLAUDE_PLUGIN_ROOT}/scripts/marketresearch/customers.mjs customers.csv --json work.json ``` A ready-made example list is in examples/sample-customers.csv if the user wants to try it before uploading their own. The parser maps the customer columns, reports how many cells are filled vs blank, and writes a per-row research worklist. Then follow references/cu